Josh Day

You may have to roll a few things on your own. Also, I've been meaning to work on StatsBase-like weights for OnlineStats so maybe this will nudge me to do...

Thanks! Your stuff could either go here or in a new package that depends on OnlineStatsBase (if you want to use the OnlineStats interface). I'm happy to have it live...

Sure! That looks very interesting. I'll give the paper a closer look, but I'm not sure if I'll have the bandwidth to work on implementing it for a while.

> How did you imagine this should work? Right now I have a fit!(MergingDigest, Vector{

> In fact, anything that supports a quantile operation can be sampled. A package that implements this would be pretty cool. e.g. ```Julia x = QuantileSampler(thing_with_quantile_method) rand(x) ``` But for...

The biggest hurdle is that an `OnlineStats.Weight` is a fundamentally different thing than `StatsBase.AbstractWeights`. I'll need to think through it more, but I probably won't have time this week.

I've thought about having `Base.empty!` methods that reset the state. I'm hesitant to add `empty!` as a required method to the OnlineStatsBase interface, but it would be nice to have...

`Base.empty!(::OnlineStat)` is an error, but I've added it as an optional part of the interface (via the OnlineStatsBase README). If anyone wants to knock out a few methods, PRs welcome.

I haven't done anything MCMC in a while, but I think OnlineStats has all the pieces you need (means, variances, and autocorrelations). The implementations of `Mean` and `Variance` live in...

Yes, either that or to have an existing Kalman filter package depend on OnlineStatsBase. I'm interested in adding it here, but I probably won't have the bandwidth to work on...